LENGTH OF THE INTERNODES 29 



As to the constancy of the number of internodes on a nerve 

 fiber, I may quote first the view of Boycott from his original 

 paper. Boj^cott says (p. 377) : 



It appears from the figures that there is a small increase in the num- 

 ber of internodes during the growth in the length of the nerve. The 

 increase is small, yet it is regularly progressive. It may be due to ordi- 

 nary errors of experiment, while on the other hand it may represent an 

 increase which actually takes place. There is one special circumstance 

 which renders it suspicious. This is the fact that the longer an internode 

 the less likely is it to remain unbroken and capable of being traced 

 throughout its length : the longer internodes will in this way not be meas- 

 ured as often as they should, and hence the average internodal length will 

 be smaller than it should be proportionately to the increase in the length 

 of the internodes, This would account for the increase which is seen in 

 the table, but whether it is the whole explanation it is impossible to 

 say. Assuming that the figures are in the main correct, it must be 

 concluded that there is a small (and somewhat doubtful) increase 

 in the number of internodes, though the main part of the total increase 

 in length is due to an increase in the length of individual internodes. 



In the summary he says again (p. 380) : 



The number of internodes thus remains approximately constant at all 

 ages. There is a small increase in the total observed number; there are 

 however reasons for thinking that this is due to errors of method. 



This question of constancy^ in the number of internodes in 

 the nerve cannot easily be decided until we have a strong reason 

 to believe that the average value obtained for the internode from 

 the given nerve is nearly correct. There is still another theoret- 

 ical objection to considering the observed values as the average 

 of the whole population, when a small fraction of the lower end 

 of the nerve only is examined (see p. 40). I have also tested this 

 main point by using mean values obtained by the present bio- 

 metric method, and the following are the results. 



